SMPS losses question- ringing

I have a question about losses on a SMPS I'm developing. This is a 2000W full-bridge SMPS for an audio amp- hard-switched at 100kHz.

If I drive the FET gates full on as fast as I can, I get this large voltage ringing on the output secondary diodes that comes close, if not exceeding, the voltage rating(600V) of the diodes. I can snub most of that ringing with parallel RC's across the diodes, but the R's need to be quite large to dissipate the amount of ringing.

I also had luck slowing the FET turn on dramatically enough to eliminate the ringing on the secondary diodes, but of course at the cost of greater FET turn on losses. I REALLY had to slow them down- enough that I run out of pulse width because of the delay created by slowing them down. I don't think this is the best way to go about it.

Any ideas on which method is better, or ideally, any ideas on how to alleviate the ringing all together? I've tried a few different transformer designs to try to lower the leakage inductance as low as possible. (Split primary, etc.) Any other primary transformer winding tricks that could help? Do I just stick with heavy sunbber losses?

Any help would be greatly appreciated
Matt. (Sorry for the long post...)

What is the SMPS topology? It would be helpful if you
would post or email the schematic. Often it is as simple
as component selection.
<als> (fsmyth_at_txucom_dot_net)

Hello

The reason of this is that simply your 100 Khz switching frequency, from my experience, switching frequencies above 30 Khz are more than moderate enough.

Regards